About the role
The Planning team decides what the vehicle does next: which gap to take, when to yield, how to pass a double-parked van without frightening the cyclist beside it. You will work on behaviour and trajectory planning that is safe by construction, smooth enough to be comfortable, and legible enough that other road users can anticipate it.
Other drivers break rules, perception is uncertain and prediction is multimodal, so the planner reasons under uncertainty rather than around a single forecast. Planning consumes tracks from Perception, poses from Localisation and constraints from Safety, and hands a trajectory to Controls. The group shares interaction-modelling ideas with Frontier AI researchers working on sequence models and reinforcement learning.
Every planner change is evaluated in closed-loop simulation against thousands of scenarios drawn from real fleet events before it is allowed near a vehicle.
What you’ll do
- Design and implement trajectory generation and selection, using optimisation, sampling or search, that respects kinematic, dynamic and comfort limits.
- Build behaviour planning for interactive situations such as unprotected turns, merges and negotiating narrow streets with oncoming traffic.
- Integrate probabilistic predictions into planning so the vehicle hedges against multiple plausible futures instead of the most likely one.
- Encode safety constraints as formal checks, such as responsibility-style rules and time-to-collision envelopes, that hold independently of the learned components.
- Analyse closed-loop simulation and fleet disengagements to find where the planner is too timid, too aggressive or simply unclear.
- Keep planner runtime inside a fixed budget, with deterministic worst-case behaviour under load.
What you’ll bring
- Experience shipping motion planning for robots or vehicles, including trajectory optimisation, lattice planners, MPC or search-based methods.
- Strong C++ engineering practice, with an eye for numerical robustness, timing and testability.
- Understanding of vehicle kinematics, collision checking over swept volumes and how planning interacts with tracking controllers.
- Familiarity with decision-making under uncertainty, such as POMDP approximations, scenario trees or risk-aware cost functions.
- A track record of tuning behaviour with data and metrics rather than by anecdote.
- Comfort reading a planner trace line by line to explain why the vehicle did what it did.
Nice to have
- Experience with learned planners, imitation learning or reinforcement learning combined with a rule-based safety layer.
- Work on game-theoretic or interaction-aware planning with other agents.
- Experience with convex optimisation solvers and warm-starting for real-time use.
- Background in human factors and what makes a vehicle's intent legible to pedestrians and drivers.
Work you might take on
- 01
Rework the lane-change decision to reason over several predicted futures for the vehicle behind, and cut hard-brake events in dense merges.
- 02
Implement a smooth stop-and-go policy for congested lanes that bounds jerk while keeping a safe, constant-time headway.
- 03
Add a formal runtime check that vetoes any trajectory violating a minimum gap to vulnerable road users, and test it adversarially.
- 04
Build a scenario miner that turns near-miss fleet logs into parameterised simulation cases for planner regression.
Compensation
$220K – $360K
The US base salary range for this role. Every offer also includes equity in Clexerl and, when it spins out, in the venture you help build. We tell you the reasoning behind your number.
Benefits
Health, fully covered
Medical, dental and vision for you and your dependants, premiums paid in full.
Two kinds of equity
Equity in Clexerl, plus a stake in the venture you help build when it spins out.
22 weeks parental leave
Fully paid, for every parent, however your family arrives.
Rest that sticks
Flexible time off with a four-week minimum we actually check on, plus a company-wide week in December.
Learning and compute
A yearly budget for courses, conferences and books, and compute credits for side projects.
401(k) with a 4% match
Matched from your first paycheque and vested immediately.
How we hire
Apply 5 minutes
Send links to what you have built and a few lines on why this role. No cover letter needed.
Intro call 30 minutes
A conversation with the hiring manager about the work, the team and what you want next.
Work sample 2–4 hours
A paid take-home or a pairing session on a real-shaped problem. You pick the format.
Studio day One day
Meet the team, go deep on your work and ours. Remote or in person.
Offer Within 5 days
A written offer with the full range, the equity and the reasoning behind both.
Opens an email with a short template. Links beat résumés.